Psychometric properties of an instrument to measure resilience in adults

A rationale for using a projective approach, in addition to self-reports, i s presented. A resilience exercise is described, consisting of 6 sentences describing adverse situations, in response to which participants write proj ective stories. A scoring scheme for such stories is introduced. 152 adults (A(age) = 34.28, SD = 9.15; M-educ = 14.55, SD = 2.31) working in organiza tions, completed the exercise and self-report scales. On the basis of initi al scoring by two judges, the scoring scheme was revised to clarify some in structions. On a new sample of 20 protocols a 0.87 agreement between two ju dges was obtained. One judge then re-scored all protocols on the revised ma nual. A word count per protocol correlated 0.54 (p < 0.000) with the total score. Scores per story and scores per scoring category, were corrected for word count, using a regression procedure. The 6 stories all loaded on a si ngle resilience factor. Exploratory and confirmatory factor analyses showed a e-factor model to fit the data best, producing factors which measured ab stract and concrete aspects. The total resilience score correlated 0.26 (p < 0.001) with Antonovsky's Sense of Coherence scale (short form) and 0.21 ( p < 0.01) with Diener's Satisfaction with Life scale.